UC Berkeley’s Doudna Lab Redesigns Minimal CRISPR Nucleases with AI, Revolutionizing In Vivo Delivery

July 26, 2026

CRISPR Medicine News USA

Overview

Researchers at UC Berkeley’s Innovative Genomics Institute (IGI), including Jennifer Doudna, have published a Science paper detailing an AI-powered technique to minimize and redesign CRISPR nucleases. This breakthrough dramatically reduces the size of gene-editing proteins, enabling their packaging into adeno-associated virus (AAV) vectors, and significantly advancing targeted in vivo delivery. This is a pivotal development in overcoming one of the biggest hurdles in in vivo gene therapy.

In Depth

Key Findings

Researchers at the Innovative Genomics Institute (IGI) at UC Berkeley, spearheaded by contributions from Nobel Laureate Jennifer Doudna, have unveiled a groundbreaking approach published in Science on July 16, 2026. Their work details the use of artificial intelligence (AI) to redesign minimal CRISPR nucleases, dramatically reducing their size. This technological leap enables efficient packaging of these gene-editing proteins into adeno-associated virus (AAV) vectors, thereby overcoming a major limitation in targeted in vivo gene therapy delivery.

Technical and Clinical Details

The research focused on optimizing some of the smallest naturally occurring CRISPR nucleases, such as Cas12f. While Cas12f already possesses an advantage over the larger Cas9 in terms of AAV packaging, full optimization for in vivo delivery remained a challenge. The AI algorithms analyzed millions of protein sequences and structural data to design variants of Cas12f that maintained their core function while achieving further miniaturization and enhanced efficiency for optimal delivery. This innovation demonstrates the potential to deliver gene-editing tools more effectively and safely to target cells without exceeding the AAV vector’s stringent payload capacity. The technology is also anticipated to contribute to reducing off-target effects and improving tissue-specific targeting capabilities.

Background & Context

In gene therapy, the delivery system, ensuring safe and efficient transport of therapeutic genes or gene-editing tools to target cells, stands as one of the most critical development challenges. AAV vectors are considered highly promising due to their safety and tissue tropism but are subject to strict limitations on the size of the genetic payload they can carry. Earlier genome editing tools like CRISPR-Cas9 have been constrained in their in vivo applications by these size restrictions. The AI-driven redesign of nucleases offers a potential solution to this delivery bottleneck, promising to dramatically expand the in vivo applicability of gene therapies.

Strategic Significance & Outlook

The AI-optimized minimal CRISPR nucleases are expected to open new avenues for treating various genetic disorders, neurodegenerative diseases, and cancers. Accelerated preclinical evaluations focusing on the efficacy, safety, and long-term stability of these redesigned enzymes in vivo are anticipated. Furthermore, this AI-driven design platform could be extended to optimize other gene-editing tools beyond CRISPR systems, as well as various therapeutic proteins, potentially having a broad impact across biopharmaceutical development. As a crucial technological innovation for the widespread adoption of gene therapy, its future research and development trajectory will be closely watched by the scientific community and industry alike.

Source: https://crisprmedicinenews.com/news/ai-redesigns-minimal-crispr-nucleases/

Get our weekly technology intelligence — free

Receive an infographic that lets you judge at a glance whether each field’s analysis report is worth reading.

Subscribe Free — Weekly Tech Intelligence

By subscribing, you’ll receive Troy-Technical’s weekly technology intelligence newsletter.